g77 error handling on sequential unformatted files

Dave Love d.love@dl.ac.uk
Wed Jan 13 05:41:00 GMT 1999


>>>>> "VRA" == Anand, Vaidyanathan R <Vaidyanathan.Anand@ARZ.Boeing.com> writes:

 VRA> Problem:
 VRA> ------------
 VRA> I/O error handling on Fortran unformatted sequential files appears to be
 VRA> incorrect; specifically, BACKSPACE after encountering a READ error is not
 VRA> done correctly. 

Incorrect in what way?  I.e. exactly how is g77's behaviour not
standard-conforming or not according to its doc?  See section 12.6 of
the f77 standard, for instance, and similar wording for f95.



More information about the Gcc-bugs mailing list